feedback-coach

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates entirely within the context of text transformation. It defines a clear workflow for a 'Feedback Coach' persona to assist law professors in improving the tone and rubric-alignment of their feedback without utilizing any dangerous tools or system permissions.
  • [SAFE]: No signs of prompt injection, data exfiltration, or obfuscation were found. The instructions are focused on pedagogical refinement and accessible communication standards.
  • [SAFE]: Analysis of indirect prompt injection surfaces confirms a low-risk profile despite processing untrusted user data.
  • Ingestion points: User-provided rubric, draft feedback, and student response (referenced in SKILL.md Step 0).
  • Boundary markers: The skill does not define specific delimiters for user data.
  • Capability inventory: The skill does not use any tools, shell commands, or network-enabled functions.
  • Sanitization: No automated sanitization of external text is described, relying on the agent's internal analysis steps.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 01:10 PM
Security Audit — agent-trust-hub — feedback-coach